The WHO has estimated that more than 36 million adults and children have been infected with HIV-1 worldwide. In sub-Saharan Africa, the AIDS pandemic has been most severe, with more than 25 million individuals - almost 10% of the total population aged 15-49 years - living with HIV-1. In Botswana, a country with 1.67 million individuals, HIV prevalence is 38.5%. A variety of biological and behavioral hypotheses have been proposed to explain this high HIV prevalence. Of particular concern is the lack of information on the role of alcohol use and HIV risk behaviors, despite anecdotal reports which suggest that alcohol may be important in HIV-related risk behaviors. The goal of this study is to determine the role of alcohol use and unsafe sex among individuals at risk for HIV-1 infection. Specifically, we plan to address the following aims: 1. To measure sexual behaviors that are associated with HIV-1 infection among a population at risk forHIV infection in Botswana. Specifically, what is the frequency of partner change, condom and alcoholuse, and what are their impact on HIV infection? 2. To explore the relationship between drinking and unsafe sex, particularly the role of alcohol andcondom use. Are certain drinking behaviors (binge drinking, alcohol dependency) associated with unsafe sex? 3. To determine the role of contextual factors such as meeting places, (i.e. bars), settings for sex, and the role of alcohol in the use of condoms. Cross-over and case crossover designs, nested within aprospective cohort, will be employed to evaluate these factors. 4. To assess feasibility and acceptability of future alcohol and HIV risk reduction interventions. Thepotential role of HIV prevention interventions that focus on alcohol risk will be assessed. If alcohol plays an important role in the sexual transmission of HIV in Botswana, this proposal will be essential in design and conduct of future HIV risk reduction interventions in Sub-Saharan Africa.